268 
BANKING STANDARDS 
It will be recalled that the district ratios in Parts II and III 
were secured by dividing one statistical aggregate by another 
and multiplying by 100. That is, for instance, the average 
ratio of gross earnings to earning assets for a district and for 
all districts was obtained by dividing the total gross earnings 
by the total earning assets and multiplying by 100. In the case 
of the ratios for the individual banks, this process was carried 
out for each bank, the resulting ratios only being made available. 
Accordingly, in order to secure.averages for all banks and for 
groups of banks, it was necessary to add and average the ratios 
themselves. By this process each bank ratio is given the same 
weight, the size of the bank, with respect to the factors deter- 
mining the ratios, being ignored. 
Marked as are the uniformities, tendencies, and correlations 
for member banks by districts and years, as shown in the pre- 
ceding pages, it is not self-evident that similar conditions char- 
acterize the interrelations of individual banks. Data which could 
be analyzed to determine these and other facts for member banks 
in the First and Second districts were fortunately available. By 
using methods of analysis similar to or identical with those 
already found to be useful in isolating and measuring these 
“master facts behind the unit facts,” the operating history, in 
the matter of gross and net earnings, in one, and of total expense 
in both districts, has been traced out in detail. The questions, 
among others, to which answers were sought are the following: 
Do member banks in their interrelations truly constitute a sys- 
tem, and if so what are its characteristics? Do the norms, trends, 
and correlations which characterize the banking structure for 
member banks by districts also characterize the individual banks 
within districts? 
The following analysis makes use of techniques suited to 
answer these and other questions. Aggregates, averages, and so 
on, are obviously required, inasmuch as it is from them only 
that generalizations are possible. In the matter now to be dis- 
cussed, they refer not to all member banks by districts but to 
individual banks by groups within districts. Moreover, inas- 
much as the sizes of the cities in which the banks are located, 
as well as their average earning assets and identity by number 
are given, it is possible to classify the data fully and to observe 
not only the norms, trends, and correlations for the group as